| Summary: | The central undertaking in this thesis is to explore the explanatory power of the concept of policy networks. The main question is whether there is a relation between the structural features of policy networks and their performance? Does network structure matter for network performance, and in that case, in what sense? In order to investigate the relationship between structure and performance, five implementation networks, engaged in inter-organizational collaboration with the task to create multidisciplinary units, at Luleå University of Technology (LTU), are studied. Each network is analyzed regarding both structural properties and performances. First, network performance is measured by the level of effectiveness and innovation. Next, the structural features of the implementation networks are measured. Drawing upon previous work of Burt, the structural analysis is based on the examination of two specific network mechanisms, namely network closure and global structural holes. Basically, while the former refers to the degree of interconnectedness, the latter considers the extent to which the actors span global structural holes, meaning that they have contacts reaching outside the network in focus. A positive relation between the two above mentioned mechanisms and performance is proposed. The empirical analysis confirms the assumption that there is a relation between structure and performance. While the existence of global structural holes is a necessity for innovative networks to form, their level of effectiveness is positively related to the degree of network closure. Following this, an innovative network is a network in which the actors are tightly connected and, at the same time, have many connections to other actors, engaged in other network constellations. Further, on the basis of the empirical findings, two new hypotheses, specifying the relationship between structure and performance, are suggested. Firstly, it is proposed that the function of prioritizing, so vital for the process of organizing, is ...
|
|---|